Natural Killer (NK) cells play a major role in defense against pathogenic infections and tumors. The missing- self hypothesis that provided a mechanistic framework for NK cell tolerance for more than 20 years has been challenged directly by recent experiments. Thus, a mechanistic understanding of signal integration from a variety of stimulatory and inhibitory receptors leading to NK cell activation and tolerance is lacking. This proposal seeks to produce a new mechanistic framework for understanding NK cell activation by synergistically combining computational approaches rooted in statistical physics, nonlinear dynamics and engineering with wet lab experiments. Genetically engineered human NK cell lines stimulated by mouse cytomegalovirus (CMV) encoded ligands will be used to validate the computational model and test model predictions. This unique experimental system allows us to quantitatively test the computational model with experimental outcomes by precisely controlling the numbers and types of the ligands and receptors. We investigate two overlapping specific aims in the proposal. 1. Determine mechanisms underlying the early time interplay between signals initiated by opposing NK receptors that lead to NK cell activation. We will develop spatially homogeneous computational model describing stochastic kinetics of early time signaling events in NK cells interacting with diverse stimulatory and inhibitory ligands. Model predictions will be tested against wet lab experiments using human NK cell lines. 2. Determine the role of spatial kinetics and clustering of receptors and associated signaling molecules in modulating stimulatory and inhibitory signals leading to NK cell activation. Spatially inhomogeneous stochastic and deterministic simulations combined with imaging and biochemical wet lab experiments using human NK cell lines will be carried out to determine the effect of microscopic clustering and spatial kinetics (diffusion and transport) of receptor-ligand complexes and receptor-bound signaling molecules. At successful completion of the proposed projects, we will have a computational model with predictive capabilities for NK cell activation that will very likely provide strategies for therapeutic interventions against viral infections (such as CMV) and tumors, as well as, create an improved mechanistic framework alternate to the existing missing-self hypothesis.